First Steps Toward Supervised Learning for Underactuated Bipedal Robot Locomotion, with Outdoor Experiments on the Wave Field
نویسندگان
چکیده
Supervised learning is used to build a control policy for robust, dynamic walking of an underactuated bipedal robot. The training and testing sets consist of controllers based on a full dynamic model, virtual constraints, and parameter optimization to meet torque limits, friction cone, and environmental conditions. The controllers are designed to induce periodic walking gaits at various speeds, both forward and backward, and for various constant ground slopes, flat ground, uphill, and downhill. They are also designed to induce aperiodic gaits that transition among a subset of the periodic gaits in a fixed number of steps. In experiments, the learned policy allows a 3D bipedal robot to recover from a significant kick. It also enables the robot to walk down a 22 degree slope and walk on sinusoidally varying terrain, all without using a camera.
منابع مشابه
Exciting passive dynamics in a versatile bipedal robot
Dynamic bipedal robots, which are capable of versatile behaviour, are rare. Most bipedal robots are either versatile and static or dynamic and limited to specific gaits. Over the last twenty years the bio-inspired spring mass model became a versatile template for bipedal walking and running gaits with many studies motivated by its high potential for versatile robot locomotion. However the chara...
متن کاملVirtual Constraints and Hybrid Zero Dynamics for Realizing Underactuated Bipedal Locomotion
Underactuation is ubiquitous in human locomotion and should be ubiquitous in bipedal robotic locomotion as well. This chapter presents a coherent theory for the design of feedback controllers that achieve stable walking gaits in underactuated bipedal robots. Two fundamental tools are introduced, virtual constraints and hybrid zero dynamics. Virtual constraints are relations on the state variabl...
متن کاملSpring-mass Walking with Atrias in 3d: Robust Gait Control Spanning Zero to 4.3 Kph on a Heavily Underactuated Bipedal Robot
We present a reduced-order approach for robust, dynamic, and efficient bipedal locomotion control, culminating in 3D balancing and walking with ATRIAS, a heavily underactuated legged robot. These results are a development toward solving a number of enduring challenges in bipedal locomotion: achieving robust 3D gaits at various speeds and transitioning between them, all while minimally draining ...
متن کاملCombining Trajectory Optimization, Supervised Machine Learning, and Model Structure for Mitigating the Curse of Dimensionality in the Control of Bipedal Robots
To overcome the obstructions imposed by high-dimensional bipedal models, we embed a stable walking motion in an attractive low-dimensional surface of the system’s state space. The process begins with trajectory optimization to design an open-loop periodic walking motion of the high-dimensional model and then adding to this solution, a carefully selected set of additional open-loop trajectories ...
متن کاملApplications of Switching Control in Robot Locomotion Workshop on Intelligent Control in Robotics and Automation
In this paper we discuss the application of switching control to problems in intelligent robot control. We will show a connection between research issues in robot locomotion and issues in robot gymnastics, and robot air hockey. These problems are fundamentally related to the control of redundant and underactuated robots. We discuss the general problem of switching control and relate these conce...
متن کامل